Title: Particle swarm optimisation with adaptive neighbourhood search for solving multi-objective optimisation problems

Authors: Layak Ali; Samrat L. Sabat

Addresses: School of Engineering, Central University of Karnataka, Gulbarga, Karnataka – 585311, India ' Center for Advanced Studies in Electronics Science and Technology, School of Physics, University of Hyderabad, Hyderabad – 500046, India

Abstract: This paper proposes a new variant of PSO algorithm with adaptive neighbourhood search of Pareto solutions to solve multi-objective optimisation problems. This algorithm essentially consists of identifying the particles responsible for worst solution and accelerating these particles towards the best solution in an adaptive manner. This concept is hybridised with the updation strategy of non-dominated solution using crowding distance. Performance of the proposed algorithm is demonstrated by solving standard multi-objective benchmark problems including CEC 2009. In particular we compare our method with two recent algorithms by considering the inverted generational distance, spacing, convergence and diversity as the performance metrics. The proposed algorithm is also compared with recent state of art methods in terms of inverted generational distance. The results demonstrate that the proposed algorithm is quite competitive in terms of the mentioned performance indicators as compared to other algorithms.

Keywords: multi-objective optimisation; Pareto optimality; particle swarm optimisation; PSO; adaptive neighbourhood search.

DOI: 10.1504/IJSI.2016.077420

International Journal of Swarm Intelligence, 2016 Vol.2 No.1, pp.1 - 42

Received: 09 Apr 2014
Accepted: 13 Dec 2014

Published online: 30 Jun 2016 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article